The hsp90 chaperones participate in the late stage maturation of proteins involved in diverse cellular activities ranging from signaling to bacterial recognition. Cytoplasmic Hsp90 guides the maturation of steroid hormone receptors, proto-oncogenic kinases, G-proteins, and other key mediators of neoplastic transformation. Inhibitors of Hsp90 that disrupt this maturation display potent anti-cancer activity. GRP94, the ER hsp90 paralog, chaperones proteins destined for transport to the cell surface and secretion, plays a key role in the export of toxic malfolded proteins from the ER, and when targeted leads to the death of tumor cells. The activity of hsp90 chaperones is regulated by the binding of ATP and other ligands to the N-terminal domain, and the response of this domain is central to the hsp90 chaperone cycle. Yet despite their high degree of sequence and structural homology, the cytoplasmic and ER hsp90 paralogs exhibit fundamental differences in their response to regulatory ligands. Ligand dependent conformational changes in the N-terminal domain lid of cytoplasmic Hsp90 have been difficult to demonstrate, and a plausible trigger for any changes has not been identified. In GRP94, on the other hand, a conserved insertion in the lid primes this region to undergo conformational rearrangements that differ dramatically depending on the identity of the incoming ligand. These rearrangements alter the quaternary structure of the chaperone. The research in this proposal will identify the ligand dependent conformational states available to GRP94 using X-ray crystallography to visualize the structures of ligand-driven GRP94 complexes. In vivo assays will be used to assess the biological importance of these structural rearrangements. The sensitive response of the GRP94 lid to incoming ligands allows GRP94 to bind inhibitors that cannot be accommodated in cytoplasmic Hsp90 and are therefore selective for GRP94. Because selective and non-selective ligands are closely related, we will use X-ray crystallography to visualize the both classes of ligands bound to GRP94 and Hsp90 and infer their mechanism of selectivity. Because each of the hsp90 paralogs is responsible for chaperoning distinct sets of client proteins, specific targeting of one hsp90 paralog with selective inhibitors may result in higher efficacy and therapeutic control. Finally, new developments offer the prospect of characterizing novel client and co-chaperone interactions with both Hsp90 and GRP94: minimal client protein loading systems open the way to isolating client-Hsp90 complexes for biochemical and biophysical study, and the recent identification of the first GRP94 accessory factor, os9, has, for the first time, opened the door to characterizing how protein-protein interactions occur in the ER paralog.